Analgesia and sedation practice in children in intensive care in Australia and New Zealand: A survey Authors: Johanna Holschier, Andrea Veysey, Kate Masterson, Debbie Long, Simon Erickson, Michele Cree, Corrine Balit, Warwick Butt, Ben Gelbart Published: 2026 ABSTRACT AbstractObjectivesTo describe the contemporary practice of analgesia and sedation in children in intensive care units (ICU) in Australia and New Zealand (ANZ).DesignA unit level survey.SettingICUs in Australia and New Zealand between December 2024 and May 2025.ParticipantsICUs that admit more than 150 children annually and report data to the Australian and New Zealand Paediatric Intensive Care Registry.Main outcome measuresProtocols for analgesia, sedation, delirium and iatrogenic withdrawal syndrome (IWS) as well as medication preferences and quality initiatives related to analgesia and sedation.ResultsFourteen of the 16 eligible ICUs completed the survey. Analgesia and sedation protocols were reported by seven tertiary paediatric ICUs (7/9) compared to two general units (adult/paediatric and combined neonatal/paediatric ICUs) (2/5). The preferred first line analgesic was morphine, and first line sedative was dexmedetomidine, however reported sedative preferences varied. Protocolised pain assessment tools were universally reported, sedation assessments were reported in 13/14 units, iatrogenic withdrawal assessments in 12/14 units and less than half (6/14) reported protocolised delirium assessment. Protocols for neuromuscular blocking agent use were also reported in 6/14 ICUs. Multidisciplinary quality assurance groups were infrequent (4/14 ICUs).ConclusionsPaediatric analgesia and sedation protocols were reported in nearly two-thirds of ICUs in ANZ that care for children and more commonly in tertiary PICUs. Analgesia, sedation and drug withdrawal protocolisation were common however, less so for delirium. Despite the multiple components of analgesia and sedation management, governance and quality assurance programs were uncommon.Keywords: Analgesia, Sedation, Paediatric intensive care unit, Delirium, Withdrawal syndrome Introduction Analgesia and sedation are fundamental components of the management of children in intensive care. The key objectives for providing analgesia and sedation are to optimise comfort and tolerance of therapies balanced against minimising complications of medications such as drug withdrawal, immobility, delirium and delayed recovery. Therefore, structured approaches to analgesia and sedation care are employed to ensure a balance between these objectives. The Paediatric Analgesia and Sedation in the Paediatric intensive care Environment (PANDEM) guidelines recommend structured and evidence-based approaches to analgesia and sedation to enable earlier mobilisation and rehabilitation from critical illness.1 They recommend protocolised pharmacotherapy, routine utilisation of validated paediatric bedside monitoring tools and incorporation of non-pharmacological interventions. There are limited data regarding such practices in children in Australia and New Zealand (ANZ). Moreover, little is known about the variability of drug preferences, assessment tools and quality initiatives. Several international surveys conducted in Europe, Canada, Japan, and most recently the United Kingdom have described preferred drug selection and dosing, monitoring tools and the availability and the use of structured protocols.[2], [3], [4], [5] Findings from these surveys suggest the need to harmonise protocols and to integrate evidence-based guidelines. A survey conducted 20 years ago in tertiary paediatric intensive care units (ICU)s in ANZ, focusing on children requiring prolonged ventilation (>72 h), revealed variability in use of sedation assessment tools and infrequency of unit protocols.6 Specific analgesia and sedation assessment tools, delirium and iatrogenic withdrawal syndrome (IWS) were not reported. Therefore, we conducted a unit level survey to describe contemporary practice in ICUs that care for children in ANZ. The objectives were to describe the use of protocols, pharmaceutical agents, assessment tools and quality initiatives such as sedation/analgesia ward rounds, governance procedures or case reviews. Method Study design A unit-level cross sectional electronic survey was distributed between December 2024 and May 2025, to ICUs that submitted data to the Australian and New Zealand Paediatric Intensive Care Registry (ANZPICR).7 The study was endorsed by the Australian and New Zealand Intensive Care Society Paediatric Study Group (ANZICS PSG) and ethics approval was provided by the Royal Children's Hospital Human Research Ethics Committee (MCC DERP/3667). Inclusion criteria Tertiary paediatric ICUs, general ICUs that admit children and paediatric-neonatal units that report at least 150 paediatric admissions per year to the ANZPICR were eligible. There were no exclusions. This study was conducted and is reported according to the Consensus-Based Checklist for Reporting of Survey Studies (CROSS)8 (Supplementary Appendix). Survey development The survey was developed by paediatric intensive care clinicians (medical, pharmacy and nursing) and informed by previous studies.[2], [3], [4], [5], [6] The aim of the survey was to describe unit-based practice and not individual practice. Domains included – 1. analgesia and sedation protocols and assessment tools, 2. drug selection, 3. delirium and drug withdrawal management, and 4. quality initiatives (Supplementary Appendix). A pilot of the survey was conducted with three clinicians (intensivist, nurse and pharmacist) to ensure content validity, clarity, readability and logical flow. The final survey was formatted in the REDCap web application, hosted by the Murdoch Children's Research Institute. The survey was distributed to ICU heads of department who delegated the most appropriate clinician to complete the survey. Questions included multiple-choice and free-text responses. The survey only allowed one response per site. No individual personal data was recorded, with only the site name recorded to ensure completion. Statistical analysis Survey responses were summarised using descriptive statistics. Categorical data were presented as frequencies and percentages. Data were displayed graphically to illustrate response distributions. Free text responses were reviewed by a single investigator (HH) and grouped into themes. All analyses were performed using Microsoft Excel. We described subgroups based on status as tertiary paediatric or general ICUs. Ethics approval and consent to participate Permission to conduct the study was provided by the Royal Children's Hospital Human Research Ethics Committee MCC DERP/3667. Results Characteristics of intensive care units Sixteen eligible units were invited to participate, of which fourteen (88%) completed the survey. Nine ICUs 9/14 (64%) were tertiary PICUs and 5/14 (36%) were general ICUs comprising one paediatric/neonatal and four paediatric/adult ICUs. The number of paediatric beds per unit is shown in Supplementary Fig. 1. Two general ICUs did not participate. Frequency of analgesia and sedation protocols The reported use of protocols for paediatric analgesia and sedation, based on ICU type, is shown in Fig. 1. Analgesia and sedation protocols were numerically more frequent in tertiary PICUs compared to general units (Tertiary 7/9 vs general 2/5). The two tertiary ICUs that did not have an analgesia and sedation protocol, had scoring tools for pain, agitation, delirium but only one had an IWS tool. The three general ICUs that did not have an analgesia and sedation protocol each had a pain and IWS tool and two had an agitation and delirium scoring tool. Frequency of analgesia and sedation protocols based on ICU type. PICU, paediatric intensive care unit. Analgesia and sedation pharmacotherapy Reported drug preferences for continuous infusion are outlined in Fig. 2. Morphine was the most commonly reported analgesic infusion. There greater uniformity in reported preferences for analgesic agents and variability for sedative agents. For sedation, 5 ICUs reported age or disease based drug preferences, whereas the most frequent, individual reported agent was dexmedetomidine (n = 4). All ICUs reported recommended agents for analgesia and sedation and 4/14 described population-specific recommendations based on age, expected duration of sedation and required depth of sedation. Morphine was the most commonly reported drug for bolus analgesic therapy (10/14) and propofol was the most common bolus drug for sedation (4/14) (Supplementary Figs. 2 and 3). Reported preferences for bolus and infusion analgesic tended to align (morphine most prefered) whereas, for sedative agents, bolus and infusion reported preferences were more varied (Supplementary Figs. 2 and 3). Reported first and second-line agents for continuous infusion of analgesia and sedation. Bedside assessment tools for pain, agitation, delirium and iatrogenic withdrawal syndrome A total of 16 assessment tools were reported for the assessment of pain, agitation, delirium and IWS. Pain and agitation All ICUs reported using an assessment tool for pain and 13/14 for agitation. A total of seven tools were reported for pain, with Face, Legs, Activity, Cry, and Consolability (FLACC) the most frequently reported, and four for sedation, with Richmond Agitation Sedation Score (RASS) the most frequently reported (Fig. 3). The most frequently reported time intervals for pain and agitation assessment were 4–6 hourly (Supplementary Figs. 4–5), twice daily for delirium (Supplementary Appendix Fig. 6) and both twice daily and 6–8 hourly for IWS (Supplementary Appendix Fig. 7). Reported assessment tools for pain, sedation, delirium and iatrogenic drug withdrawal. FLACC; Face, Legs, Activity, Cry and Consolability score, NRS; Nurse Reported Score, NIPS; Neonatal Infant Pain Score, RASS; Richmond Agitation Sedation Score, SBS; State Behavioural Scale, CAP-D; Cornell Assessment of Paediatric Delirium, p-CAMS-ICU; Paediatric Confusion Assessment Method for the Intensive Care Unit, WAT-1; Withdrawal Assessment Tool for Paediatric Withdrawal, NAS; Neonatal Abstinence Score, SOS; Sophia Observation Withdrawal Symptoms. (n = number of responding units); 14 units reported a total of 67 preferences for scoring tools across at least 15 different tools across the four domains. Delirium Protocolised delirium screening for all children was reported in 6/14 ICUs reporting two assessment tools. An additional 4/14 ICUs reported delirium screening targeted to “at-risk” children. Reported delirium assessment tools are shown in Fig. 3. Five ICUs recommended olanzapine, four quetiapine, two clonidine, one droperidol and one risperidone for delirium pharmacotherapy. Four units included more than one agent in their protocol. Iatrogenic withdrawal syndrome Protocolised recommendations for IWS treatment were reported in 13/14 ICUs. A recommended IWS assessment tool was reported by 12/13 (Fig. 3). Eight ICUs reported a specified duration of pharmacotherapy exposure prior to IWS assessment, whereas five ICUs relied on clinician discretion. Specific treatment of IWS was reported by 12/14 ICUs; seven ICUs reported protocolised choice and dose of breakthrough agents, while five applied clinician discretion. Neuromuscular blocking agents (NMBA) Recommendations for neuromuscular blocking agents (NMBAs) were reported in 6/14 ICUs, all by continuous infusion (Supplementary Appendix Fig. 8). Two ICUs reported prescribed indications for NMBAs and four ICUs reported clinician discretion (Supplementary Appendix Fig. 8). Monitoring of the depth of neuromuscular blockade was reported by five units and included both peripheral nerve stimulation and clinical assessment. Quality assurance Four ICUs reported having a multidisciplinary analgesia and sedation special interest group comprising medical, nursing and pharmacy clinicians. Two of these units included other allied health clinicians. Three units reported undertaking continuous quality improvement initiatives. The use of protocols and assessment tools based on the presence of a special interest group is shown in Supplementary Table 1. Discussion Key findings In this contemporary survey of paediatric analgesia and sedation practice in ICUs in ANZ, we found that prescribed protocols were more frequently reported in tertiary PICUs compared to general ICUs. Protocol use was reported by approximately two-thirds of ICUs. All surveyed sites reported recommendations for first- and second-line analgesic continuous infusions (predominantly opiates). Recommended drugs for sedative infusion varied but dexmedetomidine was the most frequently reported individual medication. The reporting of assessment tools was high, particularly for pain, agitation and IWS, however less so for delirium. Protocolisation for NMBA agents was reported by less than half of sites. Quality assurance processes were not commonly reported, suggesting that governance and pharmacological stewardship may be ad-hoc or not systematically developed. Comparison to other studies The implementation of dedicated analgesia and sedation protocols are recommended by paediatric consensus guidelines.1 A multi-centre North American cluster randomised trial in 2015 reported that protocolised care (including targeted sedation, arousal assessment and extubation assessment), compared to usual care, showed no overall difference in the duration of mechanical ventilation but, in exploratory analyses, reported some patient-centred benefits such as more opiate-free days, reduced exposure to sedative classes and greater comfort.1,9 Despite this result, we found that, in general protocol use was high. We found that sedative agents varied with an overall preference for alpha agonists whereas opiates were more commonly consistent across units. The transition from midazolam in ANZ is evident from the results of the prior Australian-based survey in 2008, which reported midazolam as the sedative of choice, albeit in fewer surveyed ICUs (7/8 units, 87.5%).6 The frequency of recommendation of propofol for bolus sedation was surprising given the common teaching of avoiding its use in children. Two recent clinician-based surveys, one conducted in the UK and the other within the European Society of Paediatric and Neonatal Intensive Care (ESPNIC), reported a preference for midazolam compared to alpha-agonists2,5 with only 11% preferring dexmedetomidine as first line in the ESPNIC study, demonstrating regional variability. Over the past two decades, pain and agitation assessment tools for children have increased, with less reliance on vital signs alone.1,6 We found that assessment tools were universal for analgesia and reported in all but one site for sedation and that there were a total of sixteen reported assessment tools across all domains. The most frequently reported tools, such as the COMFORT B and FLACC, were similar to the results of the European survey.2 In our survey, delirium assessment was reported both, for all children and less frequently, in targeted, at-risk groups. The reported protocolised assessment of delirium appears infrequent internationally.3,4 A National Japanese survey reported that 5/24 PICUs used targeted screening, and a Canadian survey also reported infrequent use, with 78% (n = 78) of clinicians stating they do not use any delirium screening tool.3,4 Delirium screening was not reported in the UK or ESPNIC surveys.2,5 We also reported ICU selection of a wide range of pharmacological agents for delirium, but not the indication, and recognise that there are insufficient data regarding their use in children in ICU.[10], [11], [12] The reported variation between survey results likely reflects differences in their design as well as regional variation in practice. Protocolised continuous infusion of NMBAs was reported in less than half of surveyed ICUs in our study. The Japanese and ESPNIC surveys2,4 also reported their use but not the UK or Canadian surveys.3,5 Continuous infusion of NMBAs was reported in almost all Japanese sites (21/24)4 and for specific indications in 86% (184/215) of respondents within the ESPNIC survey.2 Sites within our survey reported use of nerve stimulators for depth of blockade assessment whereas vital signs were more common in the ESPNIC survey.2 This variability likely reflects differences in approaches to avoiding neuromuscular blockade for concern over acquired neuromuscular weakness and preferring it as a targeted approach to managing mobility and avoiding sedative agents to achieve this goal. Implications of findings Our findings of high reported rates of protocolisation and the number of assessment tools indicate the inherent complexity of managing comfort in children in intensive care. These data provide important information regarding the contemporary analgesia, sedation and delirium landscape and to inform the design of prospective studies in children in ICU. Pharmacological management of comfort requires a balance between appropriate therapy and minimising complications which requires stewardship, particularly in the era of promoting early mobilisation and rehabilitation.1 In 2020, Baby SPICE, a pilot randomised trial of dexmedetomidine compared to unprescribed standard care in PICUs in ANZ, reported that dexmedetomidine was well tolerated and achieved early light sedation compared to standard care.13 It is unclear whether the results of Baby SPICE influenced the reported practice favouring alpha agonists as preferred sedative agents. A larger trial of dexmedetomidine, compared to midazolam, is underway (mini-MENDS ClinicalTrials.gov ID (NCT04801589)). Universal protocolisation of care is unlikely to be the ultimate approach to improved management of comfort in critically ill children. However, more precise objectives and targets, consistency of care and improved stewardship are likely to play a part. Strengths and limitations The strength of this survey is its application to the ANZICS network of ICUs that care for children. We sought responses at a unit level, rather than individual preferences, to describe unit-level practice and governance. Broad domains of analgesia, sedation and their complications were described. The response rate was high, however, the overall number of units was small and therefore, more detailed comparisons were not performed. Our results provide an overview of self-reported recommended protocols and practice in ANZ and do not reflect their adherence or application, nor variability of practice within ICUs. We did not aim to describe practice for subgroups of children (such as mechanically ventilated children or specific conditions such as congenital heart disease) but aimed to understand overarching approaches. Nor did we describe the adherence to protocols, which may be an important factor in the quality of analgesia and sedation delivery. We did not explore the interaction of anaesthesia-driven and ICU-driven protocols for post-operative ICU admissions and acknowledge that this may influence protocols and practice. Conclusions Analgesia and sedation protocols in ANZ were reported in nearly two-thirds of ICUs that care for children, more commonly in tertiary PICUs. Greater variability was reported for sedative agents compared to analgesics and assessment tools for most domains were common. Despite the multiple components of analgesia and sedation management, governance and quality assurance programs were uncommon. Prospective studies relating to analgesia and sedation in children in intensive care may require some harmonisation of practice. These data provide contemporary information regarding practice variability for paediatric analgesia and sedation in ICUs in ANZ. CRediT authorship contribution statement BG HH and AV were responsible for the concept, study design, data collection, data analysis and manuscript preparation; DL, SE, WB, CB, MC, KM assisted with the study design and contributed to the writing of the manuscript. Funding None declared. Conflict of interest Warwick Butt declares they are part of the CC&R editorial team as associate editors If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article. Ben Gelbart declares being a Member of the editorial board of CC&R and declares no financial or personal conflicts of interest. Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.ccrj.2026.100196. Appendix A. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 References Associated Data Supplementary Materials Multimedia component 1